mcpField
M
mcp-vscode-template

mcp-vscode-template

Local
github.com/timsonnerDeveloper Tools9Updated 80w ago
Repository

About mcp-vscode-template

mcp-vscode-template is a starter template for building an MCP server for the VS Code Agent. It provides setup guidance and a sample Python server workflow using uv.

Key features

  1. 01VS Code Agent server template
  2. 02Python project setup guidance
  3. 03uv-based environment setup
  4. 04Example MCP tool decorator

Use cases

  1. 01Start a VS Code Agent MCP server project
  2. 02Set up a Python MCP environment with uv
  3. 03Configure an MCP server in VS Code

MCP support

The template guides users through creating and running an MCP server for the VS Code Agent, including VS Code settings configuration.

#mcp#vscode#template

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

mcp-vscode-template

MCP server template for VS Code Agent

Setup

Install uv however you like. May options available.
https://docs.astral.sh/uv/getting-started/installation/

Project setup is heavily based off of Renae Schilg's work. I didn't even bother to change the project name as an homage although I did deviate on a few things, namely not using Claude Desktop, but also modified external-recon.py fairly heavily.
https://nae-bo.medium.com/building-your-first-offensive-security-mcp-server-dd655e258d5f

# Initialize project
uv init external-recon
cd external-recon

I had to modify the python versions in .python-version to 3.11 or something above 3.8 or 3.10
I also had to modify the line requires-python = ">=3.11" in pyroject.toml to something above 3.8 or 3.10 Mileage will vary... It may not be necessary.

# Create virtual environment and activate it
uv venv --python 3.11
source .venv/bin/activate

# Install mcp
uv add "mcp[cli]"

# Install dnspython (dependency for external-recon.py, not necessary for all projects, but for this one yes)
uv pip install dnspython

# Create MCP server external-recon.py file or empty and rename main.py
touch external-recon.py

The VS Code settings.json should be modified.
Use which uv to find the path to uv.
The "/path/to/project/external-recon" should refer to the project path, where the MCP server .py file is located (use absoulte path).

settings.json

    "mcp": {
        "servers": {
                "external-recon": {
                 "command": "/path/to/uv",
                 "args": [
                  "--directory",
                  "/path/to/project/external-recon",
                  "run",
                  "external-recon.py"
                 ]
                }
        }
    },

One of the main differences between Renae's work and mine is I used @mcp.tool() instead of @mcp.prompt() in external-recon.py

Start the server

From the venv of the project, start the server with uv run external-recon.py
Example:

(external-recon) user@workstation external-recon % uv run external-recon.py

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.